OpenSearch/docs/reference/analysis/analyzers/custom-analyzer.asciidoc

[[analysis-custom-analyzer]]
=== Custom Analyzer

When the built-in analyzers do not fulfill your needs, you can create a
`custom` analyzer which uses the appropriate combination of:

* zero or more <<analysis-charfilters, character filters>>
* a <<analysis-tokenizers,tokenizer>>
* zero or more <<analysis-tokenfilters,token filters>>.

[float]
=== Configuration

The `custom` analyzer accepts the following parameters:

[horizontal]
`tokenizer`::

    A built-in or customised <<analysis-tokenizers,tokenizer>>.
    (Required)

`char_filter`::

    An optional array of built-in or customised
    <<analysis-charfilters, character filters>>.

`filter`::

    An optional array of built-in or customised
    <<analysis-tokenfilters, token filters>>.

`position_increment_gap`::

    When indexing an array of text values, Elasticsearch inserts a fake "gap"
    between the last term of one value and the first term of the next value to
    ensure that a phrase query doesn't match two terms from different array
    elements.  Defaults to `100`. See <<position-increment-gap>> for more.

[float]
=== Example configuration

Here is an example that combines the following:

Character Filter::
* <<analysis-htmlstrip-charfilter,HTML Strip Character Filter>>

Tokenizer::
* <<analysis-standard-tokenizer,Standard Tokenizer>>

Token Filters::
* <<analysis-lowercase-tokenfilter,Lowercase Token Filter>>
* <<analysis-asciifolding-tokenfilter,ASCII-Folding Token Filter>>

[source,js]
--------------------------------
PUT my_index
{
  "settings": {
    "analysis": {
      "analyzer": {
        "my_custom_analyzer": {
          "type":      "custom",
          "tokenizer": "standard",
          "char_filter": [
            "html_strip"
          ],
          "filter": [
            "lowercase",
            "asciifolding"
          ]
        }
      }
    }
  }
}

GET _cluster/health?wait_for_status=yellow

POST my_index/_analyze
{
  "analyzer": "my_custom_analyzer",
  "text": "Is this <b>déjà vu</b>?"
}
--------------------------------
// CONSOLE

/////////////////////

[source,js]
----------------------------
{
  "tokens": [
    {
      "token": "is",
      "start_offset": 0,
      "end_offset": 2,
      "type": "<ALPHANUM>",
      "position": 0
    },
    {
      "token": "this",
      "start_offset": 3,
      "end_offset": 7,
      "type": "<ALPHANUM>",
      "position": 1
    },
    {
      "token": "deja",
      "start_offset": 11,
      "end_offset": 15,
      "type": "<ALPHANUM>",
      "position": 2
    },
    {
      "token": "vu",
      "start_offset": 16,
      "end_offset": 22,
      "type": "<ALPHANUM>",
      "position": 3
    }
  ]
}
----------------------------
// TESTRESPONSE

/////////////////////


The above example produces the following terms:

[source,text]
---------------------------
[ is, this, deja, vu ]
---------------------------

The previous example used tokenizer, token filters, and character filters with
their default configurations, but it is possible to create configured versions
of each and to use them in a custom analyzer.

Here is a more complicated example that combines the following:

Character Filter::
* <<analysis-mapping-charfilter,Mapping Character Filter>>, configured to replace `:)` with `_happy_` and `:(` with `_sad_`

Tokenizer::
*  <<analysis-pattern-tokenizer,Pattern Tokenizer>>, configured to split on punctuation characters

Token Filters::
* <<analysis-lowercase-tokenfilter,Lowercase Token Filter>>
* <<analysis-stop-tokenfilter,Stop Token Filter>>, configured to use the pre-defined list of English stop words


Here is an example:

[source,js]
--------------------------------------------------
PUT my_index
{
  "settings": {
    "analysis": {
      "analyzer": {
        "my_custom_analyzer": {
          "type": "custom",
          "char_filter": [
            "emoticons" <1>
          ],
          "tokenizer": "punctuation", <1>
          "filter": [
            "lowercase",
            "english_stop" <1>
          ]
        }
      },
      "tokenizer": {
        "punctuation": { <1>
          "type": "pattern",
          "pattern": "[ .,!?]"
        }
      },
      "char_filter": {
        "emoticons": { <1>
          "type": "mapping",
          "mappings": [
            ":) => _happy_",
            ":( => _sad_"
          ]
        }
      },
      "filter": {
        "english_stop": { <1>
          "type": "stop",
          "stopwords": "_english_"
        }
      }
    }
  }
}

GET _cluster/health?wait_for_status=yellow

POST my_index/_analyze
{
  "analyzer": "my_custom_analyzer",
  "text":     "I'm a :) person, and you?"
}
--------------------------------------------------
// CONSOLE

<1> The `emoticon` character filter, `punctuation` tokenizer and
    `english_stop` token filter are custom implementations which are defined
    in the same index settings.

/////////////////////

[source,js]
----------------------------
{
  "tokens": [
    {
      "token": "i'm",
      "start_offset": 0,
      "end_offset": 3,
      "type": "word",
      "position": 0
    },
    {
      "token": "_happy_",
      "start_offset": 6,
      "end_offset": 8,
      "type": "word",
      "position": 2
    },
    {
      "token": "person",
      "start_offset": 9,
      "end_offset": 15,
      "type": "word",
      "position": 3
    },
    {
      "token": "you",
      "start_offset": 21,
      "end_offset": 24,
      "type": "word",
      "position": 5
    }
  ]
}
----------------------------
// TESTRESPONSE

/////////////////////


The above example produces the following terms:

[source,text]
---------------------------
[ i'm, _happy_, person, you ]
---------------------------